Work with AI
Eight practical sessions, two evening studios and one Personal AI Work Playbook.
Use AI deliberately on real work, then keep the parts that are worth repeating.
Computer or phone with AI/GPT. Bring one small non-sensitive task, or use the supplied starter case.
Write one question, choose one task, then begin Session 1.
A checked result, a reusable method and your Personal AI Work Playbook.
THREE MOVES. THEN START.
You do not need to read the whole course first. Write one question, choose one piece of work and begin Session 1.
Write your question
Before class gets moving, record: your work or curiosity, one priority question, and what a useful result would look like.
Open my playbook →Bring yours — or use ours
Use one small email, brief, comparison, document or recurring task. No suitable task ready? Use the fictional SME quotation case.
Open starter case →Start with the job
Session 1 asks where AI genuinely helps on that job, what a useful baseline looks like and what still belongs to a person.
Begin Session 1 →Required for the live route: computer or phone with AI/GPT. No coding or deployed AI system is required.
One day. One clear step forward.
Control
Choose useful work, direct the collaboration and ground it in evidence.
Discover → Brief & steer → Context & sources
Open sessions 1–3 →Expand
Compare, communicate and challenge using the same trusted facts.
Analyse & compare → Write & communicate → Think & challenge
Open sessions 4–6 →Extend & keep
Create a usable artefact and prove the method on a second input.
Create a usable artefact → Repeat & keep
Open sessions 7–8 →GO DEEPER INTO HOW YOU WORK WITH THE MODEL.
The core course already teaches briefing and grounding. These field guides are where I keep the deeper experiments, pitfalls and current working patterns behind those skills.
Prompt Engineering
Move beyond prompt tricks. Reduce meaningful ambiguity, define outcomes and boundaries, choose interaction modes, steer deliberately and compare prompts by behavior.
Open Prompt Engineering →Context Engineering
Decide what reality AI should work from: minimum useful context, source authority, conflicts, freshness, packaging, unknowns and grounding.
Open Context Engineering →Prompt Engineering shapes the instruction. Context Engineering shapes the reality available to the model. When the job itself is handed over to AI, the next engineering problem becomes Harness and Loop Engineering inside Give AI Work.
Your three days and two nights
| When | Day 1 | Day 2 | Day 3 |
|---|---|---|---|
| 09:00–11:00 | Session 1 · Discover | Session 4 · Analyse | Session 7 · Create |
| 11:15–13:15 | Session 2 · Brief | Session 5 · Communicate | Session 8 · Repeat & final proof |
| 14:15–16:15 | Session 3 · Ground | Session 6 · Challenge | Programme completed |
| 19:30–20:30 | Night 1: Source clinic | Night 2: Receiver studio | — |
Breaks, meals and rest sit between these blocks. The host supplies dates, venue and any local start-time adjustments. The evening studios develop work you already started; no late-night technical build is needed.
Personal AI Work Playbook + one checked output or decision + one remaining limitation + one next application.
Open / save / print my playbook →
If you miss an evening studio, use its card and the next morning reconnect. A printed or paired route remains available.
FROM A PRACTICE TASK TO A REAL BUSINESS.
A camera-store specialist comparing products with AI is still doing Work With AI: choosing the job, supplying the right information, inspecting uncertainty and checking the result. ACOM shows what happens when the business wants that ability to help customers beyond one person's desk.